This is the subhead for the blog post
If you run a small business or a start-up with a limited (and probably ill-defined) advertising budget, choosing between various advertising options to get the best bang for your buck can seem very daunting. However, much to my own surprise, I have discovered it can be done, and done well! You can derive significant value even with tiny budgets– and I am talking a few hundreds of dollars here!
In this post, I’ll share five best practices that I’ve learned through my Facebook advertising experience:
1. Why Facebook?
Before getting started and going into specifics, it is important to determine whether this medium aligns well with your value proposition. Especially with time and cost constraints, you would be well served if you think clearly about this at the outset.
A few (maybe obvious) questions to ask your team would be:
– Does your target audience spend a lot of time on Facebook?
– Do you already have a Facebook page?
– Does your business have enough regular “updates” to engage its followers once you get a critical mass?
– Can you use the social element to your advantage?
Facebook may not be the right platform for you, and you should only go ahead with your campaign if the answers to the above questions justify it.
2. Know your goal(s):
This may sound clichéd, but is the most important factor in determining the success of your campaign.
What exactly are you trying to achieve through your campaign? Map out a “theory” of the intended actions by your target audience. Figure 1 shows a standard value funnel that can serve as a set of objectives. Bear in mind that your goals can be very different from the ones in this example– ranging from increasing fan interaction with your page to fueling fan growth through network effects. Whatever be the goal(s), you must have them mapped out. A good way to track actions of users beyond Facebook is to use URL tags.
3. Use your goals to determine your willingness to pay:
This is a key step to ensure you derive the maximum value from your campaign. By knowing what you want your target audience to do, you can calculate how much you value each of their actions.
Start with the lowermost section of the value funnel. A good value measure to use for paying customers is Customer Lifetime Value (CLV). Let’s assume that the CLV for a paying customer is $200. For each step, assume a reasonable conversion rate (and record your assumptions). See Figure 2 for an illustration. By multiplying the rates down the funnel, we can see that under these assumptions, the likelihood of any single impression turning into a sale is 0.00125%. This means that you would be willing to pay up to $200*0.00125% = $0.0025 per impression, or $2.5 cost-per-mille (CPM). Now, you can calculate your threshold cost-per-click (CPC), ($2.5*1000*CTR) = $2.5/click.
Similarly, you can determine the price you are willing to pay for any action a user takes. Facebook’s relatively new “Actions” feature allows you to track user actions like Page Tab Views, Page Photo Views, Page Likes, Link Clicks, Page Post Likes, Page Post Shares, Comments on Page Posts, Video Plays, Page Mentions, Check-ins and Recommendations.
When your campaign is live and you are getting enough data, this gives you great insight into how much you are paying for each “action.” A continuous comparison of performance with the defined goals and thresholds over the course of the campaign is important in employing an efficient bidding strategy.
4. Bid Smart:
Now that you know how much you are willing to pay per click, you can go ahead and bid on the higher end of Facebook’s suggested CPC range (as long as it is lower than your threshold). Once your ads start running, you can tweak your bids based on their performance. Incremental changes in bids tend to affect your ad position, and thus their performance. It is actually a virtuous cycle (Figure 3), which, if gone wrong, could also become a vicious one. By being aware of your threshold, you can guarantee that you always pay less than or equal to what you can afford to pay.
Another advantage of this virtuous cycle is that once you have a high CTR going, switching to CPM-based bidding will help reduce your overall CPC. A simplistic rule of thumb is to switch to CPM (keeping the bid constant) if your ad is performing at 0.1% CTR or higher (because at >0.1% CTR and the same bid, you get more than one click per thousand impressions, rendering each click cheaper!).
In order to determine which style of bidding is most cost-effective, it’s a good idea to set up a series of A/B tests within each campaign, comparing CPM vs. CPC bidding for identical ads, setting bids to the upper bounds of Facebook’s suggested range.
5. Actively monitor results and experiment:
All your good work can be undone if this practice is not followed. Running Facebook ads in particular, and online ads in general, is still very much an art and a science. There are no cookie-cutter solutions. Therefore, it is important to go in to the campaign with an experimental mindset, expecting to fail and learn.
Break down your budget into buckets, and set daily caps on the campaigns. Conduct lots of A/B tests by running ads with different parameters like the image, ad creative, type of ad, etc., and narrow down to best-performing ads over time. Of course, it is imperative to document all changes.
Use the rich data and reports at your disposal. The available reports – Advertising Performance, Responder Demographics, Actions by Impression Time, Inline Interactions and News Feed – are valuable. Export them as CSV files and play around with the data to derive key insights. For example, see how the responder demographic data in Figure 4 shows that females in the age group 25-34 are more than twice as likely to click on the ad in question as those in the 13-17 bracket. You can use this information to refine your targeting and test assumptions made in Figure 2.
As part of the experimentation, try to leverage Facebook’s unique value proposition. What Facebook ads lose because of the lack of user intent, they make up in the targeting capabilities. Facebook has abundant demographic data given to it by its users. Similarly, social ads like sponsored stories (that look less like an ad and more like part of the news feed) and mobile ads are generating significantly higher CTRs than regular Facebook ads.
The bottom line is that there is no ‘one-size-fits-all’ approach with Facebook advertising. You must be willing to actively monitor your campaign’s performance and tweak parameters till you get it right. As can be seen from the post, there is a fair bit of analytical work that can be put behind the advertising effort. At the very least, the above best practices will ensure that your money is well spent – and put you on course for positive outcomes!
– Last fall, as a graduate student at Stanford University, Adeel Arif took a course in Computational Advertising and learned the nuances of Facebook advertising through experimentation with a $400 budget. He then ran the first Facebook advertising campaign for a Brazilian start-up, Ingresse,that led to an increase of 15% in the number of page “Likes” over a period of just four weeks (having spent just over $100!). Adeel holds an MS in Management Science & Engineering ’12 from Stanford University, and a BS in Mechanical Engineering from BITS, Pilani- Dubai ’10.